How much would it cost to cover the uninsured in Minnesota?
Preliminary Estimates
Minnesota Department of Health, Health Economics Program
July 2006
Executive Summary
This background paper, prepared by the Health Economics Program at the request
of the Governor’s Health Cabinet, provides initial estimates of the cost to cover the
approximately 383,000 Minnesotans who lack health insurance. It uses two simplified
examples of different policy approaches to illustrate the magnitude of the potential cost to
the state to cover the uninsured population. This estimate should not be used to assess
the potential costs of any specific policy proposal.
In an approach that would use the existing MinnesotaCare program to cover the
uninsured, the total potential cost to the state is estimated at $663 million. Under an
alternative approach that would use sliding scale subsidies for the purchase of private
coverage, the estimated cost is $852 million but could be less if the state were to structure
this policy in a manner that would qualify for federal matching funds. Because the two
proposals vary in terms of the degree of subsidies provided and the benefits covered, it is
not possible to directly compare these cost estimates to determine whether a public
coverage or private coverage approach is more cost effective.
There are several important caveats to the analysis presented in this paper:
• First, this analysis is not an analysis of any specific proposal that has been made.
It is intended as an illustration of two broad approaches (public coverage vs
subsidies for private coverage) to covering the uninsured, and a starting point for
discussion about alternatives.
• Second, this analysis is intended to provide broad guidance on the cost of
covering the uninsured in Minnesota. It uses simplified assumptions and
examines general approaches and, as a result, the estimates carry a significant
degree of uncertainty. More precise cost estimates would require defining
additional details of the policy approach, creating a set of detailed assumptions
about how the behavior of individuals and employers would change in response to
the new policy, and using a complex economic modeling approach that takes into
account the multitude of factors that would interact to determine the net effect of
the policy change.
• It is unlikely that all Minnesotans would be covered by health insurance under
either of these proposals without some sort of individual mandate to purchase
coverage. It is beyond the scope of this paper to estimate any administrative costs
associated with setting up and administering systems to monitor and enforce a
mandate.
• Finally, the cost estimates in this analysis represent an analysis of potential costs
that would be paid for by the State of Minnesota. It is important to recognize
1
that this is different from estimating the overall impact on health care costs, for
two reasons:
o First, both of the proposals would involve significant shifts in the way
health care is financed. For example, over $250 million per year in
uncompensated care at Minnesota hospitals and clinics would become
insured spending.
o Second, changing the status quo would likely have effects beyond the
population that is currently uninsured. For example, low-income
individuals who purchase private coverage could drop that coverage in
order to be eligible for a subsidy, or employers could drop coverage or
reduce their contributions to premiums. These effects are not considered
in this analysis, but would be important in determining the overall impact
of any proposal.
In other words, the separate question of the overall impact on health care costs
and the distribution of costs paid by various financing sources that would
occur as a result of covering the uninsured is beyond the scope of this paper.
Introduction
This background paper has been prepared by the Health Economics Program at
the request of the Governor’s Health Cabinet. Its purpose is to provide initial estimates
of the potential cost of covering all Minnesotans who lack health insurance. It has been
prepared as a means of providing broad guidance on the cost to cover the uninsured under
different approaches, and to serve as a potential starting point for more detailed
discussion about the likely impact of specific proposals. It is important to keep in mind
that there is significant uncertainty associated with these estimates. First, this analysis
does not model a specific policy proposal but rather models the potential costs associated
two overall approaches. Second, because this analysis is intended to provide broad
guidance on the cost to cover the uninsured, this stage of analysis does not attempt to take
into account the complex interactions and factors that would be associated with analysis
of more specific proposals.
The paper discusses the potential impact of two policy options for covering the
uninsured: a “public coverage” option that would enroll the uninsured in MinnesotaCare,
and a “private coverage” option that would provide subsidies for the purchase of private
health insurance. The details that we have used in preparing these estimates were chosen
to represent two different approaches and do not represent specific policy
recommendations or proposals that have been made.
Limitations and Assumptions
The uninsured population is constantly changing. Over time, some people who
are currently uninsured will gain coverage, while others who are currently insured will
lose it. In this paper, we use estimates of the average number of uninsured in Minnesota
at a given point in time for the purpose of estimating the cost to cover the uninsured. We
estimate the cost to cover the uninsured, while assuming that people who currently have
coverage maintain their coverage. In other words, this paper essentially looks at the
2
question: if we covered all the uninsured in Minnesota tomorrow, what would the
estimated cost be under two different approaches. This paper does not incorporate any
estimates of additional cost that could be incurred due to changes in coverage among
people who are currently insured (e.g., people who are currently insured who could
qualify for a subsidy under either of the proposals analyzed in this paper).
The estimates in this paper should be considered initial estimates that are subject
to significant uncertainty. Preparing more precise estimates of how much it would cost to
cover the uninsured would require the use of sophisticated modeling techniques and very
specific sets of assumptions about the details of the policy changes that would be
proposed to accomplish this goal. Such an exercise would also require specific
assumptions about how individuals and employers would react to different policy
changes, how responsive individual and employer behavior would be to changes in
prices, and a host of other factors that would interact to determine the net effect of a
policy change.
Who Are the Uninsured in Minnesota?
The number of people who currently lack health insurance in Minnesota is
estimated at approximately 383,000.1 At any given time, more than half (61 percent) of
the uninsured in Minnesota are “long-term” uninsured – that is, people who have been
without health insurance coverage for a year or longer.2 Many of the rest of the
uninsured are likely people who are experiencing transitions in coverage – for example,
people who are in between jobs, people who have applied for public coverage and are
waiting for an eligibility determination, or people who have fallen off of public insurance
coverage for some reason (some of whom may re-enroll within a short period of time). A
significant percentage of the uninsured in Minnesota are eligible for public insurance
programs but not enrolled – for example, an estimated 59 percent of uninsured
Minnesotans are potentially eligible for public programs (based on their income) but are
not enrolled. Finally, some of the uninsured are people who could likely afford to
purchase insurance but choose not to do so; for example, an estimated 10 percent of the
uninsured (38,000 people) have incomes above 400 percent of federal poverty guidelines.
Appendix Table 1 provides more information on the characteristics of uninsured
Minnesotans, compared to the population as a whole.
Of the estimated 383,000 uninsured in Minnesota, approximately 234,000 are
long-term uninsured and the remaining 149,000 have been uninsured for less than a year.
This distinction is important for estimating the cost of covering the uninsured, because
previous research has found that the increase in health care spending that would be
predicted to occur as a result of covering the uninsured is likely to vary depending on
how long a person has been uninsured.3
1
In 2004, a survey conducted by the Minnesota Department of Health and the University of Minnesota
found that 7.4 percent of Minnesotans were uninsured at the time of the survey. This estimate of 383,000
accounts for population growth since 2004 and assumes that the uninsurance rate has remained stable since
2004.
2
Minnesota Department of Health and University of Minnesota School of Public Health, “Health Insurance
coverage in Minnesota: Trends From 2001 to 2004,” February 2006, Exhibit 1-3.
3
Jack Hadley and John Holahan, “Covering the Uninsured: How Much Would It Cost?” Health Affairs web
exclusive, June 4, 2003.
3
It is also important to consider age in estimating the cost to cover the uninsured,
because of differences in health care spending by adults and children.4 Table 1 below
shows the estimated number of uninsured adults and children in Minnesota, by length of
time uninsured.
Children (under18)
Adults
Total
Table 1 Minnesota’s Uninsured Population Uninsured longer
Uninsured less than
than 1 year
1 year
32,000
34,000
202,000
115,000
234,000
149,000
Total
66,000
317,000
383,000
Table 2 below illustrates some of the key characteristics of the uninsured,
compared to people with private health insurance and people who are enrolled in stateadministered public programs (Medical Assistance, MinnesotaCare, and General
Assistance Medical Care). Compared to people with private insurance, uninsured
Minnesotans are more likely to report being in fair or poor health. In addition, there is a
higher concentration of adults ages 18 to 34 among uninsured Minnesotans than among
the privately insured population. Compared to Minnesotans with coverage through state
programs, the uninsured are less likely to report being in fair or poor health and are more
likely to be adults between the ages of 18 and 54.
Table 2
Health Status and Age Distribution of Minnesota’s Non-Elderly Population, by Type of Insurance Coverage Uninsured
Private
Insurance
State Public
Programs*
Total
Non-Elderly
Population
Excellent
29.2%
44.3%
31.5%
43.0%
Very Good
28.7%
31.7%
27.3%
31.4%
Good
28.3%
17.2%
22.2%
18.1%
Fair
11.0%
5.0%
12.1%
5.5%
Poor
2.7%
1.9%
7.0%
1.9%
100.0%
100.0%
100.0%
100.0%
0 to 5
7.2%
7.9%
17.9%
8.8%
6 to17
10.9%
19.5%
28.4%
19.4%
18 to 24
26.0%
10.1%
13.0%
11.6%
25 to 34
23.4%
14.6%
12.8%
15.1%
35 to 54
28.2%
36.9%
21.9%
34.6%
55 to 64
4.3%
11.0%
5.9%
10.5%
Health Status
Age
4
Ibid.
4
100.0%
100.0%
100.0%
100.0%
Source: 2004 Minnesota Health Access Survey
* Minnesotans enrolled in MA, GAMC and MinnesotaCare
Public Coverage Option
One way to estimate the cost of covering the uninsured would be to estimate the
cost to cover all of the uninsured (or all of the uninsured with incomes in a specific
range) in existing public insurance programs such as MinnesotaCare. For example, the
Department of Human Services currently pays health plans a specific monthly capitation
rate for MinnesotaCare enrollees, with separate rates for children, parents, and adults
without children.5 Table 3 below shows the estimated cost of covering all of Minnesota’s
uninsured population in MinnesotaCare, assuming that DHS would pay the same rates for
the uninsured that they pay for their currently enrolled population.
Table 3 Public Coverage Option Children (under 21)
Parents
Adults w/o children
Total
State Share
Federal Share
Enrollee Premium
Number of
uninsured
94,000
153,000
136,000
383,000
Annual cost per
person*
$2,246
$4,600
$5,317
$4,277
Total cost
($ millions)
$211
$704
$723
$1,638
$663
$360
$615
*Assumes the full MinnesotaCare benefit set is provided to all adults without children (current law
provides for more limited benefits for adults with incomes between 75% and 175% of federal poverty
guidelines).
Of the approximately $1.6 billion in cost to cover the uninsured that is shown in
Table 3, not all of the money to pay for coverage would come from state funding. If
current MinnesotaCare premium schedules remained in place, about $615 million of the
$1.6 billion total would come from enrollee premiums. (The $615 million estimate
assumes that uninsured children and parents with incomes above 275 percent of poverty
guidelines and adults without children with incomes above 175 percent of poverty
guidelines would be allowed to enroll in MinnesotaCare, but would have to pay the full
premium – i.e., would not receive any public subsidy.) In addition, the state would
receive an estimated $360 million in matching federal funds for the people newly
enrolled in MinnesotaCare. The estimated net cost to the state would be $663 million.
5
Minnesota Department of Human Services
5
Because of differences in the characteristics of the uninsured versus public
program enrollees, it is important to recognize that there is uncertainty associated with
this cost estimate. For example, it is possible that the newly insured population would be
less expensive to cover than the existing population, because they have better health
status on average than the publicly insured population. On the other hand, they may have
“pent-up” needs for health care that are not currently being met due to their lack of
insurance coverage, resulting in initially higher costs than the costs for current enrollees.
Private Coverage Option
Another option that could be used in estimating the cost to cover the uninsured is
to assume that all of the uninsured enroll in private health insurance. In 2005, the
average health plan expenditure per privately insured person was $3,214.6 Using
spending estimates for relative spending of children and adults developed by Hadley and
Holahan, we estimated that the average spending per child was $1,772 and the average
spending per adult was $3,946.7 If the state were to fully subsidize the purchase of
private coverage for all of the uninsured, the total cost to cover the uninsured would be
about $1.4 billion. If the state required enrollees to contribute to the cost of premiums,
the net cost to the state would be reduced. For example, Table 4 illustrates enrollee and
state shares of premiums under a relatively simple sliding premium scale in which
enrollees with incomes below 150% of poverty guidelines receive a 100% premium
subsidy, those between 150% and 200% of poverty guidelines receive an 80% subsidy,
those between 200% and 250% of poverty receive a 60% subsidy, those between 250%
and 275% of poverty receive a 40% subsidy, and those above 275% of poverty receive no
subsidy.
It is unlikely, however, that the state could purchase private coverage for all of the
uninsured at a cost equal to the current average cost per privately insured person, because
of differences between the characteristics of Minnesota’s uninsured population and the
privately insured population. One factor that could reduce the cost compared to the
average privately insured person is the fact that Minnesota’s uninsured population is
significantly younger on average than the privately insured population. On the other
hand, the uninsured population is less healthy than the privately insured population – this
factor could increase the cost relative to the costs currently being incurred for the
privately insured population. Because of this variation, it is important to note that
estimates such as this are subject to significant uncertainty.
Another source of uncertainty in this estimate comes from the differences in cost
for individually-purchased insurance compared to employer-sponsored health coverage.
Because the share of Minnesota’s uninsured population that is eligible for employersponsored coverage is fairly low (around 20 percent), this estimate may underestimate the
6
Minnesota Department of Health, Health Economics Program, “Health Insurance Premiums and Cost
Drivers in Minnesota, 2005,” forthcoming. This estimate includes enrollment in a wide variety of health
plans, from plans with very comprehensive coverage to plans that require a high degree of enrollee cost
sharing.
7
Based on ratios of spending per child to spending per nonelderly person and spending per adult to
spending per nonelderly person from Hadley and Holahan, “The Cost of Care for the Uninsured: What Do
We Spend, Who Pays, and What Would Full Coverage Add to Medical Spending?” Kaiser Commission on
Medicaid and the Uninsured, May 2004.
6
cost of coverage because individually-purchased policies are typically more expensive
than employer-sponsored coverage, because of higher costs for underwriting and other
overhead expenses.8 (The $3,214 estimate used as the cost of private insurance per
person is representative of the fully-insured private market as a whole, which is
predominantly coverage provided through employers.) On the other hand, the potential
availability of employer contributions to coverage for the 20 percent of the uninsured
who have access to employer coverage would reduce the cost of the state subsidy for
these people.
Table 4
Private Coverage Option
Family income as %
of federal poverty
guidelines
Number of
uninsured
Total cost*
($ millions)
Enrollee share
($ millions)
State share
($ millions)
Children:
Below 150%
34,500
$61
$0
$61
150 to 200%
9,000
$16
$3
$13
200 to 250%
10,500
$19
$8
$11
250 to 275%
4,500
$8
$5
$3
Above 275%
7,500
$13
$13
$0
Below 150%
123,500
$487
$0
$487
150 to 200%
52,000
$205
$41
$164
200 to 250%
41,000
$162
$65
$97
250 to 275%
9,500
$38
$22
$16
Above 275%
91,000
$359
$359
$0
Below 150%
158,000
$549
$0
$549
150 to 200%
61,000
$221
$44
$177
200 to 250%
51,500
$180
$72
$108
250 to 275%
14,000
$46
$28
$18
Above 275%
98,500
$372
$372
$0
383,000
$1,368
$516
$852
Adults
Total
Total
* assumes cost of $1,772 per child and $3,946 per adult
8
One additional source of uncertainty that is beyond the scope of this paper is that if the state were to
require that uninsured individuals purchase private insurance coverage, it would likely also need to require
guaranteed issue of insurance policies in the individual market. Implementing a shift to guaranteed issue in
the individual market would also have an impact on premiums.
7
In comparing the differences between the public and private options described
above, it is important to remember that there are major differences in the benefit sets.
Private insurance policies are typically subject to a significant amount of enrollee cost
sharing such as premiums and deductibles. The state could potentially reduce its net cost
by obtaining federal matching funds for these subsidies; however, complying with federal
requirements that limit the amounts enrollees pay in premiums and out of pocket cost
sharing would potentially increase the total (state + federal) cost of the subsidies.
Individual Mandates for Health Insurance Coverage
It is unlikely that Minnesota could reduce its number of uninsured residents to
zero without some sort of mandate that everyone either purchase health insurance or
enroll in public insurance programs for which they are eligible. With most types of
mandates there is an element of noncompliance, and this would likely be the case with an
individual health insurance mandate as well. In order to be effective, therefore, any
individual mandate would likely need to include enforcement mechanisms as well as
penalties for noncompliance.
The advantages and disadvantages of an individual mandate for health insurance
coverage have been considered at length in policy debates and academic publications
over more than a decade.9 The major arguments used to support or oppose the idea of an
individual mandate are summarized below:
• Arguments in support of an individual mandate:
o It is highly unlikely that universal coverage could be achieved without
an individual mandate.
o An individual mandate would bring everyone into the insurance risk
pool, ensuring that people who purchase insurance are not forced to
cross-subsidize the cost of uncompensated care provided to those can
afford insurance coverage but choose not to purchase it.
o An individual mandate is “fair” in the sense that everyone is expected
to contribute to the system in accordance with his or her financial
ability.
• Arguments in opposition to an individual mandate:
o Because the cost of health insurance would be higher as a share of
income for poor people than for people with higher incomes, an
individual mandate would be similar to a regressive tax (unless the
mandate were combined with a system of income-related subsidies to
offset the regressive effect of the mandate).
o If policymakers choose to subsidize the purchase of coverage to offset
the cost of the mandate for low-income populations, the cost of
establishing and administering the subsidy system could be substantial.
We have not attempted to estimate these costs.
9
See, for example, C. Eugene Steuerle, “Implementing Employer and Individual Mandates,” Health
Affairs, Spring (II) 1994.
8
o The administrative costs of enforcing an individual mandate may be
substantial. These include the costs of determining compliance with
the mandate, as well as the costs of establishing and operating an
administrative mechanism for collecting penalties for noncompliance.
We have not attempted to estimate these costs.
Costs to the State vs. Overall Health Care Cost
The estimates presented above represent potential additional costs that would be
incurred by state government to cover the uninsured. However, these costs would not all
be “new” costs to the health care system – the uninsured receive health care services
today, and those services are paid for in a variety of ways. The two proposals to cover
the uninsured described above would entail substantial shifts in the way that health care
spending in Minnesota is financed.
Although detailed analyses of these shifts in financing and of the potential net
increase in overall health care spending in the state are beyond the scope of this paper, a
few key facts are worth keeping in mind:
• First, over $250 million per year is currently spent on uncompensated care at
hospitals and clinics in Minnesota. (Most, but not all, uncompensated care
costs are incurred by people who lack health insurance.) Explicitly funding
this care by providing insurance coverage would reduce the need for other
cross-subsidies that are currently built into the health care system, such as
public subsidies to offset the cost of uncompensated care (e.g., over $130
million in federal and state payments to Minnesota hospitals were made in
2004 to offset these costs) or cross-subsidies that are currently being paid by
people with private insurance who pay higher prices to cover the cost of
uncompensated care provided to the uninsured. Indeed, one of the original
premises behind the MinnesotaCare program was that a significant portion of
the program’s spending would represent a shift of funds that were already in
the health care system, as opposed to new spending.
• Extending coverage to all of the uninsured would likely increase overall
health care spending in Minnesota, because the newly insured population
would increase its use of health care relative to current usage. On average, the
increase in health care spending would be higher for adults than for children,
and higher for people who have been uninsured for 1 year or more.10 We did
not attempt to estimate the change in overall spending in Minnesota that
would result from covering all of the uninsured.
10
Jack Hadley and John Holahan, “The Cost of Care for the Uninsured: What Do We Spend, Who Pays,
and What Would Full Coverage Add to Medical Spending?”, Kaiser Commission on Medicaid and the
Uninsured, May 2004.
9
Appendix Table 1: Characteristics of Uninsured Minnesotans Compared to Population
Age
0 to 5
6 to 17
18 to 24
25 to 34
35 to 54
55 to 64
65+
Race/Ethnicity 1
White
Black
American Indian
Asian
Other Race
Hispanic/Latino
Country of Origin 2
US Born
Hispanic Nation
Asian Nation
African Nation
Other Nation
Family Income, as % of Poverty Guidelines
0-100%
101-200%
201-300%
301-400%
401%+
Education 3
Less than high school
High school graduate
Some college/tech school
College graduate
Postgraduate
Health Status
Excellent
Very Good
Good
Fair
Poor
Uninsured Minnesotans
All Minnesotans
7.2%
10.9%
25.9%
23.3%
28.0%
4.3%
0.4%
100.0%
7.8%
17.0%^
10.2%^
13.3%^
30.5%
9.2%^
12.1%^
100.0%
70.1%
7.6%
5.6%
4.8%
0.2%
15.1%
see note
88.7%^
4.4%^
2.0%^
3.6%
0.1%
3.3%^
see note
82.1%
11.9%
3.1%
1.5%
1.5%
100.0%
93.1%^
1.8%^
2.1%
1.0%
1.9%
100.0%
26.3%
31.5%
20.6%
11.9%
9.8%
100.0%
9.5%^
16.6%^
17.7%
18.9%^
37.3%^
100.0%
18.9%
36.3%
32.1%
10.7%
2.0%
100.0%
7.1%^
24.4%^
34.0%
23.0%^
11.4%^
100.0%
29.1%
28.7%
28.5%
11.0%
2.7%
100.0%
39.5%^
31.0%
19.9%^
7.1%^
2.6%
100.0%
Source: 2004 Minnesota Health Access Survey.
^ Indicates a statistically significant difference between uninsured and all Minnesotans at the 95% level
1 Distribution adds to more than 100% since individuals were allowed to choose more than one race/ethnicity
2 Country of origin is only reported for individuals 3 and older
3 Education refers to the parent's highest level of education obtained for individuals under 18
10